Modulation signature trigger
A trigger generator and trigger method are provided for determining whether or not a signal under test matches a modulation signature. The modulation signature may be provided as a magnitude signature, a phase signature or both. When the magnitude values, phase values, or both of a signal under test are the same as their respective modulation signature, an error computation will be close to zero. If this value is within a threshold value, a trigger signal or other indication of a match is produced.
Latest Tektronix, Inc. Patents:
- INTEROPERABILITY PREDICTOR USING MACHINE LEARNING AND REPOSITORY OF TX, CHANNEL, AND RX MODELS FROM MULTIPLE VENDORS
- METHODS FOR 3D TENSOR BUILDER FOR INPUT TO MACHINE LEARNING
- REAL-EQUIVALENT-TIME OSCILLOSCOPE
- ULTRA-WIDEBAND SIGNAL GENERATOR WITH LOW PHASE NOISE
- COMPREHENSIVE MACHINE LEARNING MODEL DEFINITION
Embodiments of the present invention relate to test and measurement instrumentation, and more particularly to providing a trigger signal based upon a detection criteria.
Referring now to
In response to specified conditions that define a trigger event, the trigger generator outputs a trigger signal. A memory controller is said to capture I-Q data in response to the trigger signal. In some embodiments, the act of acquiring, or capturing, I-Q data is accomplished by holding data that is already stored in the memory, such that it will not be overwritten during the normal acquisition process. In other embodiments, the I-Q data may be transferred from a temporary memory, into a more permanent memory, or other storage medium for subsequent processing. Depending upon the desired action, I-Q data from some period prior to the trigger signal, some period subsequent to the trigger signal, or a combination thereof, may be captured, or stored, in response to the trigger signal. In some embodiments, the I-Q data corresponding to the signal that met the defined trigger event are also captured and stored.
The word trigger may generally refer interchangeably to the trigger circuit, the trigger signal, and in some instances the type of signal event that results in a trigger signal. Triggers are becoming more and more sophisticated as modem communications systems have become more complex. There is a growing desire to identify a variety of signals, or signal anomalies, that occur very infrequently in signals that may be present for very long periods of time. In these systems, while a problem may be known to exist, the source of the problem may be difficult to isolate or identify. This is especially true when the problem is intermittent. Long data records, on the order of hours or even day, may have to be examined to try to find the data corresponding to an event.
As signals become more complex, simple level triggers in the time domain, and even frequency mask triggers in the frequency domain are insufficient for providing the ability to trigger on more complex signals, or signal anomalies.
SUMMARYAccordingly a test and measurement instrument is provided comprising a modulation signature trigger. The modulation signature trigger provides a means for comparing an input signal against a modulation signature, such as a magnitude signature or a phase signature. A detector compares the magnitude values, or phase values against a corresponding modulation signature. When a match is indicated, such as by an error computation being produced that is within a threshold value, a trigger signal or other indication is produced. In some embodiments, the signal under test may be captured, or acquired, based upon the trigger signal. In other embodiments, a marker may be associated with stored signal data when a match is indicated.
A real time spectrum analyzer, according to an embodiment of the present invention, has the ability to capture, or acquire, a set of I and Q data samples as the result of a triggering event. In an embodiment of the present trigger system, a trigger event can be based upon Magnitude-Phase data set. A user can provide an expected or desired waveform in the time-domain before the measurement and the instrument compares the incoming data samples and evaluates whether there is a match with the provided waveform for each sample.
When making measurements, the magnitude values from the signal under test (SUT) are not necessarily at the same scale as the magnitude signature. The signal under test can be scaled, have some constant level shift, or both, relative to the signature. Accordingly, in an embodiment of the present trigger system, the detector 58 provides a scaling factor and scales the magnitude signature so that the incoming signal and the scaled magnitude signature are matched. In one embodiment, the scaling factor is estimated to minimized the estimation error (vector e). In another embodiment, a shift level is estimated to minimize the estimation error. In a further embodiment both a scaling factor and a shift level are estimated to minimize the estimation error. In one embodiment of the present invention, a Least Squares Estimator is used to find this scaling factor. For example, the following formula can be used as a basis for minimizing the error:
where the sampled signal is vector y, the signature is vector x, the scaling factor is a, the constant level shift, or offset, is b. This error may also include thermal noise in the signal. In an alternative embodiment, the incoming signal is scaled and/or offset to match the magnitude signature.
Once a match is found, a trigger signal is generated to identify or acquire the signal. In some embodiments the signal under test is being written to an acquisition memory or other storage as I and Q signal, and the trigger signal causes this memory to be saved, rather than overwritten. In a further embodiment, the trigger signal may cause the relevant portion of the acquisition memory to be stored to a longer term storage device such as a hard drive. Depending upon the settings, the trigger signal may cause signal information from before, during, and/or after the trigger causing event to be acquired, or saved. In an embodiment of the present invention, signal information related to the portion of the signal under test that caused the trigger signal to be produced is acquired.
In another embodiment, shown in
For phase or frequency modulated signals, the modulation depth can be scaled or have some constant level shift. Furthermore, there may be a frequency offset between signal and signature. By estimating these possibly unknown parameters to minimize the estimation error a better comparison can be made with the phase signature. The following relationship can be used to estimate a scaling factor, frequency offset, and constant phase shift (or offset):
where, the sampled signal is vector y, the signature is vector x, time is vector t, scaling factor is a, frequency offset is Δf constant phase shift (or offset) is b, and the error vector is e.
As described above the signature vector x is being adjusted by estimating the appropriate scaling, frequency offset, and phase shift values. As would be understood by one skilled in the art, it would be similarly possible to estimate similar values to scale the signal vector y instead.
If the frequency deviation of a signal is known, such as in the case of Minimum Shift Keying (MSK) modulation, an alternative embodiment of the present invention provides a fixed scaling factor a reducing the number of factors that need to be estimated allowing accuracy to be obtained more efficiently. The phase and frequency offset is used to identify the phase signature from the signal under test. As shown in
In several embodiments described above in connection with both magnitude signature and phase signature, a straight line fit is computed as part of the detection. In embodiments of the present invention, a straight line is determined using a Least Square Estimation (LSE). Taking for example formulas having the same form:
y=Hθ+e
One can estimate the parameter vector, θ, by least square fit to minimize the squared error. The estimator, {circumflex over (θ)}, is which can be solved to the equation:
The resulting least squared error (LSE) is
where I is the identity matrix. Finally, we can compare the LSE to a given threshold value to determine whether to generate the trigger signal.
Another embodiment of the present invention is shown in
where W is a nonsingular symmetric matrix for weighting. It is generally, inverse of error correlation matrix, R. In this case, the diagonal elements are the squared values of the magnitude signature and other elements are zeros if we assume all error elements are independent each other. If the magnitude signature has the form,
m=[m0, m1, . . . , mN-1]T,
then the matrix W is
The estimator, , is:
where y is the phase difference vector. The LSE is obtained as:
LSE=yT[W−WH(HTWH)−1HTW]y.
The embodiments described above relate to a real-time trigger that is implemented in hardware, such as an ASIC, FPGA or other customized circuitry. By implementing this in hardware, the comparison can be made and a trigger signal generated in real time, meaning that the trigger can be produced without dropping a sample or reducing the sample rate. In an alternative embodiment, these structures and operations can be implemented in software running on a programmable processor, such as a general purpose processor, a digital signal processor, or other processor capable of running software. This allows for post process analysis to identify matches to the modulation signatures. In a further embodiment, where hardware has been provided to perform this operation in real time, the trigger generator circuit 40 is used to provide post process analysis. In this embodiment, the processor 32 provides data from an acquisition memory 36, or other storage to the trigger generator 40. If a match is found to a modulation signature, the trigger signal is used to provide a marker associated with the stored signal. This can be accomplished by having the trigger write to the memory directly as shown in
Particular aspects of a signal under test can be found, or identified, by combining both phase signatures and magnitude signatures. For example, a phase signature trigger can be combined with a magnitude trigger by ANDing each comparators trigger signals to provide a resulting trigger signal when the logic condition is met. Alternatively, a trigger could be constructed by ORing phase signature triggers, and magnitude signature triggers.
In a further embodiment, the signature triggers described above could be implemented sequentially to find complicated triggers. The first trigger could use a first method, or a first signature, and subsequent triggers could use a different method, or signature, to construct a trigger system that would be able to find a portion of a signal under test containing more complicated signatures.
Modulation signature triggers may be useful to search a symbol pattern in a symbol sequence. Recent digital modulations use phase information to carry messages. However, absolute phase and absolute timing are often unknown to a receiver. To determine the absolute phase and time, finding fixed symbol patterns is helpful. Embodiments of the present invention simplify the pattern search. Ordinary symbol by symbol pattern search needs to prepare and compare all possible phase patterns. Especially, if the modulation type is Offset Quadrature Phase-Shift Keying (OQPSK), or similar modulation type, a conventional symbol search has to look at two types of decoded symbol sequences: In-Phase first symbol sequence and Quadrature-Phase first symbol sequence. Embodiments of the present invention deal with the half symbol shift by preparing the desired symbol signature at half symbols.
Claims
1. A test and measurement instrument comprising:
- a signal input to obtain magnitude values or phase values corresponding to a signal under test;
- a detector to compare the magnitude values, or the phase values against a modulation signature; and
- a comparator to generate a trigger signal when a portion of the signal under test matches the modulation signature;
- wherein the modulation signature is a phase signature; and
- wherein the detector compares the difference between the phase values of the signal under test and the phase signature.
2. The test and measurement instrument of claim 1, wherein the detector applies weighting based upon a magnitude signature corresponding to the phase signature.
3. A method comprising the steps of:
- obtaining magnitude values or phase values for a signal under test;
- comparing the magnitude values or phase values against a modulation signature; and
- indicating that a portion of the signal under test matches the modulation signature;
- wherein the step of comparing phase values against a modulation signature further comprises subtracting a phase signature from the phase values to obtain a phase differences signal.
4. The method of claim 3, further comprising calculating a straight line fit to the phase differences.
5. The method of claim 4, further comprising providing a magnitude signature along with a phase signature and applying a weighting factor based upon the magnitude signature when calculating the straight line fit.
4096397 | June 20, 1978 | Dreiling |
4403184 | September 6, 1983 | Witt et al. |
4491930 | January 1, 1985 | Hyatt |
4686655 | August 11, 1987 | Hyatt |
4797936 | January 10, 1989 | Nakatsugawa et al. |
4914418 | April 3, 1990 | Mak et al. |
5341435 | August 23, 1994 | Corbett et al. |
5841286 | November 24, 1998 | Stoops |
7009170 | March 7, 2006 | Dobbs et al. |
7286755 | October 23, 2007 | Goodwill et al. |
7349457 | March 25, 2008 | Stoddard et al. |
20060176151 | August 10, 2006 | Ems et al. |
20070100548 | May 3, 2007 | Small |
20090082982 | March 26, 2009 | Cain |
1669764 | June 2006 | EP |
1672376 | June 2006 | EP |
1724595 | November 2006 | EP |
1826576 | August 2007 | EP |
- Sivula A Ed—Institute of Electrical and Electronics Engineers: “Next-generation devices and networks bring opportunities and challenges,” Proceedings International Test Conference 2003. (ITC). Charlotte, NC, Sep. 30-Oct. 2, 2003; [International Test Conference], New York, NY: IEEE, US, vol. 1, Sep. 30, 2003, p. 1319, XP010685369, ISBN: 978-0-7803-8106-3.
- Da Silva M: “Trigger-Capture-Analyze: Three Steps Toward Understanding Today's Complex RF Signals,” Microwave Journal, Horizon House Publications, Norwood, MA, US, vol. 47, No. 2, Mar. 1, 2004, p. 22, 24, 26, 28, 30, 32, 34, 36, XP008062558, ISSN: 0192-6225.
Type: Grant
Filed: Oct 9, 2007
Date of Patent: Oct 19, 2010
Patent Publication Number: 20090094495
Assignee: Tektronix, Inc. (Beaverton, OR)
Inventor: Shigetsune Torin (Beaverton, OR)
Primary Examiner: Thomas Valone
Attorney: Matthew D. Rabdau
Application Number: 11/869,637
International Classification: G01R 23/16 (20060101);